andygrove commented on PR #6940: URL: https://github.com/apache/datafusion-comet/pull/6940#issuecomment-6453072491
Thanks @parthchandra. I added the complex and temporal test you suggested, as `complex and temporal types are written through Hadoop` (4535d1e217). It keeps your shapes (an array of structs, a map with struct values, `DECIMAL(38, 10)`, a timestamp and a timestamp_ntz, CJK text), checks the native write plan and the read-back, and also compares against Spark's own writer. One change: the date starts at `1600-01-01` instead of `0001-01-01`. On Spark 3.5 the `0001` dates make Spark's own writer fail with `WRITE_ANCIENT_DATETIME` before they reach Comet. The test passes on 3.5, 4.0 and 4.1. The merge with main (62a41cd340) also needed a fix: #7186 made `get_execution_context` return a `Result`, so `setOutputStream` now propagates it with `?` (cc74578fd5). -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
